Journal of Intelligent Information Systems
Parameter Screening and Optimisation for ILP using Designed Experiments
The Journal of Machine Learning Research
Data-driven adaptive selection of rules quality measures for improving the rules induction algorithm
RSFDGrC'11 Proceedings of the 13th international conference on Rough sets, fuzzy sets, data mining and granular computing
Heuristic rule-based regression via dynamic reduction to classification
IJCAI'11 Proceedings of the Twenty-Second international joint conference on Artificial Intelligence - Volume Volume Two
A study of different quality evaluation functions in the cAnt-Miner(PB) classification algorithm
Proceedings of the 14th annual conference on Genetic and evolutionary computation
Improving the cAnt-MinerPB classification algorithm
ANTS'12 Proceedings of the 8th international conference on Swarm Intelligence
Rule quality measure-based induction of unordered sets of regression rules
AIMSA'12 Proceedings of the 15th international conference on Artificial Intelligence: methodology, systems, and applications
Evaluating the use of different measure functions in the predictive quality of ABC-miner
Proceedings of the 15th annual conference companion on Genetic and evolutionary computation
Editorial: Modifications of the construction and voting mechanisms of the Random Forests Algorithm
Data & Knowledge Engineering
CHIRA---Convex Hull Based Iterative Algorithm of Rules Aggregation
Fundamenta Informaticae
Redefinition of Decision Rules Based on the Importance of Elementary Conditions Evaluation
Fundamenta Informaticae
| Hi-index | 0.00 |
The primary goal of the research reported in this paper is to identify what criteria are responsible for the good performance of a heuristic rule evaluation function in a greedy top-down covering algorithm. We first argue that search heuristics for inductive rule learning algorithms typically trade off consistency and coverage, and we investigate this trade-off by determining optimal parameter settings for five different parametrized heuristics. In order to avoid biasing our study by known functional families, we also investigate the potential of using metalearning for obtaining alternative rule learning heuristics. The key results of this experimental study are not only practical default values for commonly used heuristics and a broad comparative evaluation of known and novel rule learning heuristics, but we also gain theoretical insights into factors that are responsible for a good performance. For example, we observe that consistency should be weighted more heavily than coverage, presumably because a lack of coverage can later be corrected by learning additional rules.